Immune System and Cancer
Malignant tumors are inherently resistant to conventional therapies and present significant therapeutic challenges. Immunotherapy has become an evolving area of research and an additional option for the treatment of certain types of cancers. The immunotherapy approach rests on the rationale that the immune system may be stimulated to identify tumor cells, and target them for destruction.
Numerous studies support the importance of the differential presence of immune system components in cancer progression (Jochems and Schlom, Exp Biol Med, 236(5): 567-579 (2011)). Clinical data suggest that high densities of tumor-infiltrating lymphocytes are linked to improved clinical outcome (Mlecnik et al., Cancer Metastasis Rev.; 30: 5-12, (2011)). The correlation between a robust lymphocyte infiltration and patient survival has been reported in various types of cancer, including melanoma, ovarian, head and neck, breast, urothelial, colorectal, lung, hepatocellular, gallbladder, and esophageal cancer (Angell and Galon, Current Opinion in Immunology, 25:1-7, (2013)). Tumor immune infiltrates include macrophages, dendritic cells (DC), mast cells, natural killer (NK) cells, naive and memory lymphocytes, B cells and effector T cells (T lymphocytes), primarily responsible for the recognition of antigens expressed by tumor cells and subsequent destruction of the tumor cells by T cells.
Despite presentation of antigens by cancer cells and the presence of immune cells that could potentially react against tumor cells, in many cases the immune system does not get activated or is affirmatively suppressed. Key to this phenomenon is the ability of tumors to protect themselves from immune response by coercing cells of the immune system to inhibit other cells of the immune system. Tumors develop a number of immunomodulatory mechanisms to evade antitumor immune responses. For example, tumor cells secrete immune inhibitory cytokines (such as TGF-β) or induce immune cells, such as CD4+ T regulatory cells and macrophages, in tumor lesions to secrete these cytokines. Tumors have also the ability to bias CD4+ T cells to express the regulatory phenotype. The overall result is impaired T-cell responses and induction of apoptosis or reduced anti-tumor immune capacity of CD8+ cytotoxic T cells. Additionally, tumor-associated altered expression of MEW class I on the surface of tumor cells makes them ‘invisible’ to the immune response (Garrido et al. Cancer Immunol. Immunother. 59(10), 1601-1606 (2010). Inhibition of antigen-presenting functions and dendritic cell (DC) additionally contributes to the evasion of anti-tumor immunity (Gerlini et al. Am. J. Pathol. 165(6), 1853-1863 (2004).
Moreover, the local immunosuppressive nature of the tumor microenvironment, along with immune editing, can lead to the escape of cancer cell subpopulations that do not express the target antigens. Thus, finding an approach that would promote the preservation and/or restoration of anti-tumor activities of the immune system would be of considerable therapeutic benefit.
Immune checkpoints have been implicated in the tumor-mediated downregulation of anti-tumor immunity. It has been demonstrated that T cell dysfunction occurs concurrently with an induced expression of the inhibitory receptors, CTLA-4 and programmed death 1 polypeptide (PD-1), members of the CD28 family receptors. Nevertheless, despite extensive research in recent years, the success of immunotherapy in a clinical setting has been limited. Few therapeutic agents have been approved by regulatory authorities, and among those, the benefit has been observed only in a minority of patients. In recent years, immune checkpoints have been implicated in the downregulation of anti-tumor immunity and used as therapeutic targets. Studies have shown that T cell dysfunction occurs concurrently with an induced expression of the inhibitory receptor, programmed death 1 polypeptide (PD-1). PD-1 is an inhibitory member of the CD28 family of receptors that in addition to PD-1 includes without limitation CD28, CTLA-4, ICOS and BTLA. However, to date, these approaches have met with limited success. While promise regarding the use of immunotherapy in the treatment of melanoma has been underscored by the clinical use and even regulatory approval of anti-CTLA-4 (ipilimumab) and anti-PD-1 drugs (pembrolizumab and nivolumab) the response of patients to these immunotherapies has been limited. Recent clinical trials, focused on blocking these inhibitory signals in T cells (e.g., CTLA-4, PD-1, and the ligand of PD-1 PD-L1), have shown that reversing T cell suppression is critical for successful immunotherapy (Sharma and Allison, Science 348(6230), 56-61 (2015); Topalian et al., Curr Opin Immunol. 24(2), 202-217 (2012)). These observations highlight the need for development of novel therapeutic approaches for harnessing the immune system against cancer.
Melanoma
Melanoma, one of the most deadly cancers, is the fastest growing cancer in the US and worldwide. Its incidence has increased by 50% among young Caucasian women since 1980, primarily due to excess sun exposure and the use of tanning beds. According to the American Cancer Society, approximately 76,380 people in the US will be diagnosed with melanoma and 10,130 people (or one person per hour) are expected to die of melanoma in 2016. In most cases, advanced melanoma is resistant to conventional therapies, including chemotherapy and radiation. As a result, people with metastatic melanoma have a very poor prognosis, with a life expectancy of only 6 to 10 months. The discovery that about 50% of melanomas have mutations in BRAF (a key tumor-promoting gene) opened the door for targeted therapy in this disease. Early clinical trials with BRAF inhibitors showed remarkable, but unfortunately not sustainable, responses in patients with melanomas with BRAF mutations. Therefore, alternative treatment strategies for these patients, as well as others with melanoma without BRAF mutations, are urgently needed.
Human pathological data indicate that the presence of T-cell infiltrates within melanoma lesions correlates positively with longer patient survival (Oble et al. Cancer Immun. 9, 3 (2009)). The importance of the immune system in protection against melanoma is further supported by partial success of immunotherapies, such as the immune activators IFN-α2b and IL-2 (Lacy et al. Expert Rev Dermatol 7(1):51-68 (2012)) as well as the unprecedented clinical responses of patients with metastatic melanoma to immune checkpoint therapy, including anti-CTLA-4 and anti-PD-1/PD-L1 either agent alone or in combination therapy (Sharma and Allison, Science 348(6230), 56-61 (2015); Hodi et al., NEJM 363(8), 711-723 (2010); Wolchok et al., Lancet Oncol. 11(6), 155-164 (2010); Topalian et al., NEJM 366(26), 2443-2454 (2012); Wolchok et al., NEJM 369(2), 122-133 (2013); Hamid et al., NEJM 369(2), 134-144 (2013); Tumeh et al., Nature 515(7528), 568-571 (2014). However, many patients fail to respond to immune checkpoint blockade therapy alone. The addition of virotherapy might overcome resistance to immune checkpoint blocking agents, which is supported by animal tumor models (Zamarin et al., Sci Transl Med 6(226), 2014).
Poxviruses
Poxviruses, such as engineered vaccinia viruses, are in the forefront as oncolytic therapy for metastatic cancers (Kim et al., 2009). Vaccinia viruses are large DNA viruses, which have a rapid life cycle (Moss et al., 2007). Poxviruses are well suited as vectors to express multiple transgenes in cancer cells and thus to enhance therapeutic efficacy (Breitbach et al., 2012). Preclinical studies and clinical trials have demonstrated efficacy of using oncolytic vaccinia viruses and other poxviruses for treatment of advanced cancers refractory to conventional therapy (Park et al., 2008; Kim et al., 2007; Thorne et al., 2007). Poxvirus-based oncolytic therapy has the advantage of killing cancer cells through the combination of cell lysis, apoptosis, and necrosis. It also triggers innate immune sensing pathway that facilitates the recruitment of immune cells to the tumors and the development of anti-tumor adaptive immune responses. The current oncolytic vaccinia strains in clinical trials (JX-594, for example) use wild-type vaccinia with deletion of thymidine kinase to enhance tumor selectivity, and with expression of transgenes such as granulocyte macrophage colony stimulating factor (GM-CSF) to stimulate immune responses (Breitbach et al., 2012, Curr Pharm Biotechnol). Many studies have shown however that wild-type vaccinia has immune suppressive effects on antigen presenting cells (APCs) (Engelmayer et al., 1999; Jenne et al., 2000; Deng et al., 2006; Li et al., 2005; ref from Deng et al., J VI 2006 paper) and thus adds to the immunosuppressive and immunoevasive effects of tumors themselves.
Poxviruses however are extraordinarily adept at evading and antagonizing multiple innate immune signaling pathways by encoding proteins that interdict the extracellular and intracellular components of those pathways (Seet et al. Annu. Rev. Immunol. 21377-423 (2003)). Chief among the poxvirus antagonists of intracellular innate immune signaling is the vaccinia virus duel Z-DNA and dsRNA-binding protein E3, which can inhibit the PKR and NF-κB pathways (Cheng et al. Proc. Natl. Acad. Sci. USA 894825-4829 (1992); Deng et al. J. Virol. 809977-9987 (2006)) that would otherwise be activated by vaccinia virus infection. A mutant vaccinia virus lacking the E3L gene (ΔE3L) has a restricted host range, is highly sensitive to IFN, and has greatly reduced virulence in animal models of lethal poxvirus infection (Beattie et al. Virus Genes. 1289-94 (1996); Brandt et al. Virology 333263-270 (2004)). Recent studies have shown that infection of cultured cell lines with ΔE3L virus elicits proinflammatory responses that are masked during infection with wild-type vaccinia virus (Deng et al. J. Virol. 809977-9987 (2006); Langland et al. J. Virol. 8010083-10095). The inventors have reported that infection of a mouse epidermal dendritic cell line with wild-type vaccinia virus attenuated proinflammatory responses to the TLR agonists lipopolysaccharide (LPS) and poly(I:C), an effect that was diminished by deletion of E3L. Moreover, infection of the dendritic cells with ΔE3L virus triggered NF-κB activation in the absence of exogenous agonists (Deng et al. J. Virol. 809977-9987 (2006)). The inventors of the present disclosure have also showed that whereas wild-type vaccinia virus infection of murine keratinocytes does not induce the production of proinflammatory cytokines and chemokines, infection with ΔE3L virus does induce the production of IFN-β, IL-6, CCL4 and CCL5 from murine keratinocytes, which is dependent on the cytosolic dsRNA-sensing pathway mediated by the mitochondrial antiviral signaling protein (MAVS; an adaptor for the cytosolic RNA sensors RIG-I and MDA5) and the transcription factor IRF3 (Deng et al., J Virol. 2008 November; 82(21): 10735-10746.). See also international Application PCT/US2016/019663 filed by the inventor and co-workers on Feb. 25, 2016; and provisional application No. 62/149,484 filed on Apr. 17, 2015 and its corresponding international application, PCT/US2016/028184. These applications are herein incorporated by reference in their entirety.
E3LΔ83N virus with deletion of the Z-DNA-binding domain is 1,000-fold more attenuated than wild-type vaccinia virus in an intranasal infection model (Brandt et al., 2001). E3LΔ83N also has reduced neurovirulence compared with wild-type vaccinia in an intra-cranial inoculation model (Brandt et al., 2005). A mutation within the Z-DNA binding domain of E3 (Y48A) resulting in decreased Z-DNA-binding leads to decreased neurovirulence (Kim et al., 2003). Although the N-terminal Z-DNA binding domain of E3 is important in viral pathogenesis, how it affects host innate immune sensing of vaccinia virus is not well understood. The inventors have previously shown that myxoma virus but not wild-type vaccinia infection of murine plasmacytoid dendritic cells induces type I IFN production via the TLR9/MyD88/IRF5/IRF7-dependent pathway (Dai et al., 2011). Myxoma virus E3 ortholog M029 retains the dsRNA-binding domain of E3 but lacks the Z-DNA binding domain of E3. It was found that the Z-DNA-binding domain of E3 (but probably not Z-DNA-binding activity per se) plays an important role in inhibiting poxviral sensing in murine and human pDCs (Dai et al., 2011; Cao et al., 2012).
Deletion of E3L sensitizes vaccinia virus replication to IFN inhibition in permissive RK13 cells and results in a host range phenotype, whereby ΔE3L cannot replicate in HeLa or BSC40 cells (Chang et al., 1995). The C-terminal dsRNA-binding domain of E3 is responsible for the host range effects, whereas E3LΔ83N virus with deletion of the N-terminal Z-DNA-binding domain is replication competent in HeLa and BSC40 cells (Brandt et al., 2001). Because E3LΔ83N is 1000-fold more attenuated than wild-type vaccinia, in this application, the inventors explored its use as an attenuated replication competent vaccinia viral vector for further construction of immune-stimulating immunotherapeutic agent against various cancers.
Vaccinia virus (Western Reserve strain; WR) with deletion of thymidine kinase is highly attenuated in non-dividing cells but is replicative in transformed cells (Buller et al., 1988). TK-deleted vaccinia virus selectively replicates in tumor cells in vivo (Puhlmann et al., 2000). Thorne et al. showed that compared with other vaccinia strains, WR strain has the highest burst ratio in tumor cell lines relative to normal cells (Thorne et al., 2007). The inventors selected a derivative of this strain, vaccinia E3LΔ83N WR strain as their vector for further modification.
Human Flt3L (Fms-like tyrosine kinase 3 ligand), a type I transmembrane protein that stimulates the proliferation of bone marrow cells, was cloned for in 1994 (Lyman et al., 1994). The use of hFlt3L has been explored in various preclinical and clinical settings including stem cell mobilization in preparation for bone marrow transplantation, cancer immunotherapy such as expansion of dendritic cells, as well as an vaccine adjuvant. Recombinant human Flt3L (rhuFlt3L) has been tested in more than 500 human subjects and is bioactive, safe, and well tolerated (Fong et al., 1998; Maraskovsky et al., 2000; Shackleton et al., 2004; He et al., 2014; Anandasabapathy et al., 2015). Much progress has been recently made in the understanding of the critical role of the growth factor Flt3L in the development of DC subsets, including CD8α+/CD103+ DCs and pDCs (McKenna et al., 2000; Waskow et al., 2008; Liu et al., 2007; 2009; Naik et al., 2006; Ginhoux et al., 2009).